基于形态特征提取的急性下壁心肌梗死BiLSTM网络辅助诊断算法

急性下壁心肌梗死是一种病发急、进展快、致死率高的心脏疾病,该文提出一种新颖的基于形态特征提取的BiLSTM神经网络分类的急性下壁心肌梗死辅助诊断算法,可大幅度提高医生对急性下壁心肌梗死疾病的诊断效率并有助于及时确诊.算法包括:对胸痛中心数据库心拍信号进行降噪及心拍分割;根据临床心内科医学诊断指南提取了12导联波形距离特征和分导联波形幅值特征;依据提取的特征搭建LSTM与BiLSTM神经网络进行心拍的分类识别;使用PTB公开数据库和胸痛中心数据库多临床中心进行交叉验证.实验结果表明,加入胸痛中心真实临床数据后,基于形态特征提取BiLSTM神经网络的急性下壁心肌梗死辅助诊断算法准确率达到99.72％,精度达到99.53％,灵敏度达到100.00％,同时F1-Score达到99.76.该算法比其他现有算法准确率提高至少1％,该项研究具有非常重要的临床应用价值.     

Double emulsion (W/O/W) gel stabilised by polyglycerol polyricinoleate and calcium caseinate as mangiferin carrier: insights on formulation and stability properties

Mangiferin (MGF) is a phenolic compound isolated from mango, but its poor solubility significantly limits its use. In this study, MGF was embedded into the inner aqueous phase of W₁/O/W₂ emulsions. Firstly, the dissolution method of MGF was determined. MGF remained stable in solution with pH 13 at 30 min, and its solubility reached 10 mg mL⁻¹. When the pH of MGF solutions was adjusted from pH 13 to pH 6, MGF did not immediately crystallise, providing sufficient time to construct the MGF-loaded W₁/O/W₂ emulsions. Subsequently, the MGF-loaded W₁/O/W₂ emulsions were constructed using polyglycerol polyricinoleate (PGPR) and calcium caseinate (CAS). The formation and stability of the W₁/O/W₂ emulsions were investigated. The MGF-loaded W₁/O/W₂ emulsions stabilised with 1% PGPR and 1% – 3% CAS exhibited a low viscosity, limited loading capacity, and poor stability. Conversely, the MGF-loaded W₁/O/W₂ emulsions stabilised by 3%PGPR–3%CAS exhibited optimal loading capacity (encapsulation efficiency = 95.31% and loading efficiency = 0.91%) and stability, which was attributed to the fact that high viscosity and gel state retarded the migration of inner aqueous phase. These results indicated that the W₁/O/W₂ emulsions stabilised by PGPR and CAS may be a potential alternative for encapsulating mangiferin.     

IRS辅助的SWIPT系统中基于能效优先的波束成形设计与优化

以智能反射面(intelligent reflecting surface,IRS)辅助的无线携能通信(simultaneous wireless information and power transfer,SWIPT)系统为背景,研究了该系统中基于能效优先的多天线发送端有源波束成形与IRS无源波束成形联合设计与优化方法。以最大化接收端的最小能效为优化目标,构造在发送端功率、接收端能量阈值、IRS相移等多约束下的非线性优化问题,用交替方向乘子法(alternating direction method of multipliers,ADMM)求解。采用Dinkelbach算法转化目标函数,通过奇异值分解(singular value decomposition,SVD)和半定松弛(semi-definite relaxation,SDR)得到发送端有源波束成形向量。采用SDR得到IRS相移矩阵与反射波束成形向量。结果表明,该系统显著降低了系统能量收集(energy harvesting,EH)接收端的能量阈值。当系统总电路功耗为?15 dBm时,所提方案的用户能效为300 KB/J。当IRS反射阵源数与发送天线数均为最大值时,系统可达最大能效。     

生物矿化法制备氧化镓颗粒及其性能表征

以蚕丝蛋白为模板,在相对温和的条件下通过生物矿化的手段形成具有特殊形貌的α-GaOOH颗粒,并通过在不同温度下煅烧α-GaOOH得到α-Ga2O3和β-Ga2O3.采用扫描电镜(SEM)、透射电镜(TEM)、X射线衍射仪(XRD)和荧光分光光度计(PL)等手段研究了丝素蛋白多肽和矿化时间对颗粒的影响,对其生物矿化机理进行了初步探讨.结果表明,所制备的β-Ga2O3具有优良的发光特性,丝素蛋白多肽模板以无定形的结构与产物结合在一起,并且经过高温烧结后仍以碳膜的形式包覆在材料的表面.这种碳膜结构对于提高材料的生物学性能起着重要的作用.     

基于发展科学探究能力的高中化学教学实践 ——以"元素周期律"章节为例

基于学生在学习元素周期律时存在的普遍问题改进了教学实践,分析了元素周期律发展学生科学探究能力的价值,设计了四个教学环节——趣味导入,成果展示;分析规律,做出猜想;设计方案,收集证据;梳理模型,迁移应用,并提出了三点针对学生探究能力发展的教学建议.     

负极添加NMP对电池性能的影响

锂离子电池负极浆料中添加N-甲基吡咯烷酮(NMP),在极片加热干燥时不能完全除去.残留的NMP吸附在石墨表面,使得SEI膜与石墨结合不牢固.研究了其对电池性能的影响,结果显示,在电池的循环过程中,SEI膜逐渐脱落,导致电池循环寿命缩短,但对电池的容量、内阻、倍率、存储等初期性能不影响.